1. Field of the Invention (Technical Field)
The present invention relates to resonant electrical circuits containing inductors and capacitors, particularly when the inductance is very low and the circuit must resonate at a low frequency. More specifically, this disclosure is directed to a method for tuning and matching extremely small sample coils with very low inductance for use in magnetic resonance experiments conducted at low frequencies.
2. Background Art
Magnetic resonance experiments and procedures typically are performed using an electrically resonant circuit, with a coil (inductor L) wrapped around a sample to be evaluated, and a capacitor (capacitance C) connected to the coil to form a series or parallel resonant LC circuit. The frequency at which the electrical resonance must occur is determined by the strength of the magnetic field used, as well as the properties of the nucleus (or electron) being studied.
The arts of Nuclear Magnetic Resonance (NMR) and Magnetic Resonance Imaging (MRI) recently have advanced in the field of sample miniaturization. Miniature sample magnetic resonance systems employ very small sample coils. Such small coils allow NMR experiments on, for example, mass-limited samples or highly localized MRI information. The use of small coils, which have low inductance, in electrically resonant circuits becomes increasingly difficult as the size of the coil is reduced and/or the required resonance frequency is reduced. The resonance frequency is typically low for miniaturized experiments, when a small, weak permanent magnet is used. To date, NMR and MRI experiments at low frequencies (below 100 MHz) have been performed with “millicoils” or “mini-coils” of relatively large size. Experiments using very small coils have always been carried out at relatively much higher frequencies. In contrast, the invention disclosed hereinafter allows the use of the smallest possible coils in NMR and MRI procedures carried out at the lowest possible frequencies.
Very small coils for use in NMR spectroscopy at relatively high frequencies have been described previously.
T. L. Peck, et al., were among the first to extend the analysis of coil circuit performance in magnetic resonance to very small coils. They identified the signal-to-noise ratio advantage of small diameter coils for samples with limited volume. A Peck, et al. study, conducted at 4.7 Tesla (200 MHz), was of small coils with diameters ranging from 0.050 mm to 0.860 mm and larger. However, even with their smallest coil, which had an inductance of only 4 nH, Peck, et al. did not face the challenge of resonating a very small coil at a very low frequency. Their tuning circuit layout discussion focused on the lengths of the leads of their small coil, although they discuss only the resistance of these leads, and not their stray inductance. Peck, et al., did not use a tuning inductor. T. L. Peck, R. L. Magin, and P. C. Lauterbur, “Design and Analysis of Microcoils for NMR Microscopy,” J. Magn. Reson. 108, 114-124 (1995).
U.S. Pat. No. 6,788,061 to Sweedler, et al., appears to describe an NMR apparatus with a sample holder having a containment region that holds a volume of less than about 1 microliters of the analyte sample, and a coil which encloses the containment region of the analyte sample holder and the contained sample. The coil is operatively associated with the analyte sample in the containment region of the sample holder, such that the coil can transmit and/or receive energy from the analyte sample in the containment region. This '061 patent contemplates operation of the system at very high frequencies, and mentions a sample of volume <10 microliters and magnets less than 50 kg. However, even though small coil usage in mico-NMR applications at low magnetic field are discussed in the '061 patent, no reference is made to resonance tuning. The '061 patent describes experiments carried out in a magnetic field of 7 Tesla, corresponding to a resonance frequency of 300 MHz. Hence, Sweedler, et al. did not confront the problem of resonating a small coil at low frequency.
U.S. Pat. No. 5,684,401 to Peck, et al., appears to teach compensation of magnetic susceptibility variation in NMR microspectroscopy detection coils. The disclosed apparatus does not employ a tuning inductor. When discussing a tuning circuit, the '401 patent concedes that their preference to move the tuning capacitance physically away from the small coil will result in a degradation of electrical performance. This concession follows convention, which teaches that extra “lead,” or “stray” inductance is to be avoided.
Seeber, et al., explored microcoil performance experimentally, using a field of 9 Tesla (383 MHz). While they used coils as small at 0.020 mm in diameter, their tuning circuits evidently never contained a tuning inductor, as their high operating frequency made this unnecessary. Seeber, et al., attempted to reduce stray inductance by placing very small capacitors as close as possible to the coils. They also made an extensive, systematic study of the deleterious effects of stray inductance. D. A. Seeber, R. L. Cooper, L. Ciobanu, and C. H. Pennington, “Design and testing of high sensitivity microreceiver coil apparatus for nuclear magnetic resonance and imaging,” Rev. Sci. Inst., 72, 2171 (2001).
U.S. Pat. No. 6,242,915 to Hurd seems to teach a field-frequency lock system for an MRI system that includes a small coil and resonant sample located to sense changes in the polarizing magnetic field. The apparatus is operated at a high frequency of 205 MHz. Changes are detected as a shift in frequency of the NMR signal produced by the resonant sample, and the frequency shift is used to compensate the MRI system. This patent deals with moderately large coils for use at moderately high frequency and does not seem to require any special tuning schemes beyond the traditional.
The prior art examples listed above suggest that the challenge of resonating a very small coil (<20 nH) at low frequencies (<100 MHz) has not been faced because the work has all been carried out in strong magnetic fields, and has involved 1H nuclei.
Moresi and Magin describe a low field NMR system using a permanent magnet operating at 0.6 T (25.5 MHz), and discuss motivations for assembling a small system using a permanent magnet. Moresi, et al., purport to that show a system can perform NMR by providing a sample coil with diameter 4 mm and inductance 168 nH. With such a relatively large coil, they faced no particular challenge in building a resonance circuit, thus employed a conventional circuit design. G. Moresi and R. L. Magin, “Miniature permanent magnet for table-top NMR,” Concept. Magn. Reson. 19B, 35-43 (2003).
Goloshefsky, et al., discuss small-coil-based NMR and MRI systems for industrial applications, and the advantages of using very small coils. Goloshefsky et al.'s system operates at a low magnetic field of 0.6 Tesla, corresponding to an operating frequency of around 25.5 MHz. However, their coils are relatively very large (two flat spirals, each with outer diameter 3.5 mm), with a large enough inductance (81 nH) that they can resonate in the traditional manner without significant difficulty. A. G. Goloshevsky, J. H. Walton, M. V. Shutov, J. S. de Ropp, S. D. Collins, M. J. McCarthy, “Development of low field nuclear magnetic resonance microcoils,” Rev. Sci. Inst. 76, 024101 (2005).
Sorli, et al., describe the design and construction of a planar small coil system for operation at 2 Tesla (85 MHz). They point out that very small coils have the potential for integration with microfluidic “lab-on-a-chip” devices. The coil of the Sorli, et al. system is rather large (0.5 mm on a side), and they describe their resonance circuit as “conventional.” B. Sorli, J. F. Chateaux, M. Pitival, H. Chahboune, B. Favre, A. Briguet, P. Morin, “Micro-spectrometer for NMR: Analysis of small quantities in vitro,” Meas. Sci. Technol. 15, 877-880 (2004).
The three immediately previous publications, concerning the use of so-called “microcoils” to perform NMR at low frequency, utilized relatively large (˜1 mm diameter, L≧100 nH) coils. Another particularly active area of application of small coils at relatively low frequency is in MRI, where the typical magnetic field strength is 1.5 Tesla, yielding a resonance frequency of about 64 MHz. The following publications are representative of efforts in this general area.
U.S. Patent App. Publication No 2005/0245814 to Anderson, et al., appears to disclose a method for determining the position and/or orientation of a catheter or other interventional access device or surgical probe using phase patterns in a magnetic resonance (MR) signal. The process employs a large coil (4 mm outside diameter), said to operate at about 1.5 Tesla (˜64 MHz). There is no apparent reference to any resonance tuning procedures for the coils.
Published International Patent Publication No. WO2005026762, to Weiss, appears to show an MR process for locating a medical instrument with a very small coil attached thereto in the examination volume of an MR device. The coil is part of a resonant circuit matched to the resonant frequency of the MR device and having no external controls. According to the disclosure, the small coil is part of a resonant circuit tuned to the resonant frequency of the MR device, which circuit is unconnected to any of the other components of the MR device. However, the does not appear to be any detailed teaching of a tuning method or of the magnetic filed used.
European Patent EP1304581 to Gleich shows a method for localizing an object, preferably a medical instrument, introduced into a body. The object is in the examination volume of an MR device, which evaluates the interaction between an electromagnetic resonant circuit, mounted on the object, and an RF field applied in the MR device for nuclear magnetization of the body. While an extremely small coil is used in the resonance circuit, no description is provided of a method or means for tuning.
International Patent Publication No. WO0173460 to Fuderer, et al., offers an interventional magnetic resonance method utilizing a very small coil. The method purports to enable localization of an interventional instrument by detection of magnetic resonance signals from the surroundings of the small coil under the influence of magnetic field gradients. The disclosure focuses on localization method through micro-NMR, but does not include any tuning references or frequency details.
Canadian Patent CA2342047 to Raghavan, et al., shows a device, such as a medical device, having a distribution of “microcoils” (pairs) that may be used within an organism under MRI visualization. At least one or each microcoil of the opposed pair of microcoils has at least a region where a diameter circumscribed by a first winding is greater than the diameter circumscribed by at least one complete second winding, especially an adjacent winding displaced from the first winding along an axis or core of the medical device or an axis of the microcoil. The second winding is nearer to or farther from an intermediate region between the microcoils that define the pair of microcoils. The device description does not include references or details regarding the frequency or the tuning procedure. The preferred coils have relatively large diameters between 1 mm and 4 mm.
U.S. Pat. No. 6,512,941 to Weiss, et al., discloses a device and method for exciting the nuclear magnetization in a limited volume of an object to be examined, utilizing a very small coil which is present in the volume and is attached, for example, to an interventional instrument during the formation of a magnetic resonance image of the object to be examined. However, no tuning of the coil is described.
U.S. Pat. No. 6,397,094 to Luedeke, et al., teaches an MR method which utilizes a very small coil without connection leads which causes a change in phase of an external RF magnetic field in its direct vicinity within an object to be examined. This increase apparently can be used to localize the coil, to image the direct vicinity, or to track the propagation of a liquid flow passing through the direct vicinity. However, no tuning of the microcoil is described.
Accordingly, the prior art apparently makes no reference to the difficulty of resonating very small coils at low frequencies. All coils with inductances comparable to 25 nH have been utilized, it is believed, at frequencies of 200 MHz and above. Coils used at 64 MHz or less have all been large, with diameters on the scale of millimeters, with inductances exceeding 80 nH. The particular challenge of resonating the smallest coils at low frequencies has not been faced, and therefore it previously has not been solved.
The challenge has also not been anticipated because those skilled in the art likely would apply the conventional solution to resonating a small coil at low frequency, that is, by simply increasing the capacitance. The problem with such an approach is that resonating a 10 nH inductor at 40 MHz, for example, would require a capacitor of value of 1580 pF. Since the electrical resonance frequency must match the magnetic resonance frequency, the capacitance must be adjustable. An adjustable capacitor of value above 1000 pF would be physically very large. An alternative is to use both a fixed-value capacitor of small physical size but large capacitance together with a physically small adjustable capacitance. The drawback of this perceived solution is a dramatic reduction in the range of capacitance adjustability.
The historical development of the art relating to the optimization of detector coil performance in magnetic resonance experiments is helpfully revealed in the seminal publication by D. I. Hoult and R. E. Richards, “The SNR of the NMR Experiment,” J. Magn. Reson. 24, 71 (1976). Hoult and Richards highlight the then-state-of-the-art understanding of coil performance (exemplified by their first reference: A. Abragam, Principles of Nuclear Magnetism, Oxford University Press, pp. 71-83 (1961)), by identifying shortcomings in the known art and proposing a more fundamental, general, and accurate approach to understanding the performance of detector coils.
The state-of-the-art prior to the 1976 Hoult and Richards publication expressed coil performance in terms of coil volume, sample filling factor (the fraction of the coil volume occupied by the sample in the usual situation where there is a single coil used as the NMR sample coil), coil inductance, and resonant circuit quality factor (“Q factor”), among other concepts. The conventional wisdom with regard to sample filling factor was to maximize the sample's “exposure” to the radiofrequency magnetic field generated by all parts of the resonance circuit. This meant, for example, minimizing the lengths of wires connecting the coil to the rest of the resonant circuit in order not to generate “wasted” fields.
Hoult and Richards developed an approach based on concepts more closely tied to the physics of the detection process. They utilized the Principle of Reciprocity to rigorously calculate the strength of the signal that would be detected, and ascribing the electrical noise that serves to obscure the desired signal to various details of the experiment. The end result of Hoult and Richards' efforts was an equation for calculating detector coil performance that is better grounded in fundamental principles than previous approaches. Specifically, their result is expressed in terms of the coil's efficiency in producing magnetic field at the sample location (defined as the field produced per unit current in the coil) and the resistance in the resonant circuit. Taking this result as a guide for designing detector coils and circuits, a researcher in the art is inexorably led to the conclusion that the best design maximizes the magnetic field intensity generated by a unit current in the coil, divided by the square root of the circuit resistance. A practical realization of this criterion is to minimize the time required to tip the spins by, say, 90 degrees, the so-called “90-degree pulse.”These design goals are equivalent to maximizing the Q-factor, inductance, filling factor, etc., only for a simple solenoid that is full of the sample and is a part of a compact resonant circuit—the usual case for NMR/MRI. In the general case, one must use the Hoult and Richards criterion, and not the older approach. This fact seems to have remained unrecognized prior to the present disclosure, even in the extension of the Hoult and Richards methodology to the regime of very small detector coils presented by Peck, et al. T. L. Peck, et al., “Design and Analysis of Microcoils for NMR Microscopy,” J. Magn. Reson. B 108, 114 (1995).
An adverse aspect of the Hoult and Richards approach is that the concepts used to describe coil performance are abstract and somewhat difficult to measure for many practitioners. The conceptual units of the previous approach of Abragam, for example, are more concrete and readily measured. Ideas like inductance, volume, filling factor, Q-factor, etc. are understood from introductory-level physics or radio technology, and there are inexpensive instruments to measure these quantities and simple methods for calculating them. Hoult and Richards recognized the conceptual value of the other approach, and showed in an appendix how their result could be used to derive the Abragam-style result, under the assumption that the filling factor was maximized. Significantly, while the Hoult and Richards result can be used to derive the Abragam-style result, the reverse is not true. The Hoult and Richards result is more general.
Thus, it is believed that the early formulation of coil performance in terms of inductance, volume, filling factor, and Q-factor have seen continued use in the art of magnetic resonance, and Abragam's book remains a popular reference in the field. The fact that the Abragam-style approach dominates current thinking is readily seen in the literature, including references mentioned hereinabove. Evidently, the Hoult and Richards result did not succeed in altering the standard methodology for designing optimal detector coils for magnetic resonance experiments, because the early formulation works for the vast majority of NMR/MRI applications. Indeed, when the Hoult and Richards article is cited in the literature, it is often used to support the notion that filling factor should be maximized, which is a misinterpretation of the result given in their appendix. From this it is evident that concepts such as the filling factor, Q-factor, etc. are highly misleading when it comes to building optimized detector coils.
The present invention was developed against the foregoing background. The present approach is contrary to prior art teaching, because the prior art has not recognized the full implications of the Hoult and Richards result. This disclosure provides a solution for tuning small inductances at low frequencies that is much more convenient than the conventional approaches. Beyond convenience, the disclosed apparatus and method permits the resonant circuit and the magnet to be made very small, which is crucial for new applications in portable MRI, for example.